wl1251: Add support for idle mode On Nokia N900 the wl1251 consumes the most power when the interface is up but not associated to access point (that supports PSM). In terms of battery current consumption, the consumption is ~180 mA higher when the interface is up but not associated and only ~5 mA higher when associated compared to interface down and driver not loaded cases. This patch adds support for the mac80211 idle notifications. Chip is put into idle very much the same way when entering into PSM by utilizing the Extreme Low Power (ELP) mode. I.e. idle is entered by setting necessary conditions in wl1251_ps_set_mode followed by a call to wl1251_ps_elp_sleep. It seems it is just enough the authorize ELP mode followed by CMD_DISCONNECT (thanks to Kalle Valo about the idea to use it). Without disconnect command the chip remains somewhat active and stays consuming ~20 mA. Idle mode is left by same way than PSM. The wl1251_join call is used to revert the CMD_DISCONNECT. Without it association to AP doesn't work when trying second time. With this patch the interface up but not associated case the battery current consumption is less than 1 mA higher compared to interface down case. Signed-off-by: Jarkko Nikula <jhnikula@gmail.com> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
wl1251: Prepare for idle mode support RFC for WL1251 idle mode support brought a few issues that are worth to update before adding the idle mode support. Since the idle mode can reuse the code that is now used in Power Save Mode (PSM), the flag psm in struct wl1251 is changed to variable station_mode to be able to distinguish between PSM and idle modes. As the station mode is different than the power power save mode command that is sent to chip, the enum wl1251_cmd_ps_mod values are used only when communicating with the chip and new enum wl1251_station_mode values are used inside the driver. Confusing comment about psm and elp relation is removed since the PSM is actually activated by putting the chip into Entreme Low Power (ELP) mode. Signed-off-by: Jarkko Nikula <jhnikula@gmail.com> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
wl1251: Send synchronization timeout event in any state Send synchronization timeout event in any state to let the firmware monitor the connection state. Signed-off-by: Janne Ylalehto <janne.ylalehto@nokia.com> Reviewed-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com> Signed-off-by: Yuri Ershov <ext-yuri.ershov@nokia.com>
wl1251: fix deadlock with ieee80211 beacon filter calls Lockdep shows that there is a deadlock when calling ieee80211_rssi_changed() under wl->mutex. Fix it by releasing mutex before calling mac80211 functions. This patch is based on Work-In-Progress patch sent by Kalle Valo. Signed-off-by: Vidhya Govindan <vidhya.govindan@nokia.com> Signed-off-by: Yuri Ershov <ext-yuri.ershov@nokia.com>
wl1251: Fix TSF calculation Cast MSB part of current TSF to unsigned long long to prevent loss of significant bits. MSB should be shifted on 32 bits. Signed-off-by: Yuri Kululin <ext-yuri.kululin@nokia.com> Signed-off-by: Yuri Ershov <ext-yuri.ershov@nokia.com>
wl1251: use ELP wakeup interrupt instead of polling The chip can generate an interrupt on ELP wakeup, make use of it instead of polling the status register. This reduces wakeup time to ~7ms from ~8-20ms (the wait loop was sleeping, so could take up to 20ms depending on how system timer is configured). Note that DEBUG_PSM does not report this accurately as it only measures jiffies. This patch also moves atomic portion of interrupt processing to main.c to avoid code duplication. Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
wl1251: cancel elp_work early While working on PS I've noticed elp_work is kicking rather often, and sometimes the chip is put to sleep before 5ms delay expires. This seems to happen because by the time wl1251_ps_elp_wakeup is called elp_work might still be pending, and if the processing takes longer (for example interrupts may take some time to process), elp_work might get started and hit the mutex. After main thread finishes work it calls wl1251_ps_elp_sleep, which can no longer cancel old work (as it's already started) and schedules yet another work needlessly. The elp_work then gets the mutex and puts the chip to sleep too early. Fix this by cancelling work in wl1251_ps_elp_wakeup instead. Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
wl1251: remove wl1251_ps_set_elp function wl1251_ps_set_elp() only does acx_sleep_auth call and takes the chip from/to ELP, however all callers of wl1251_ps_set_mode() have already taken the chip out of ELP and puts it back to ELP when they finish. This makes ELP calls (and register writes they result in) superfluous. So remove wl1251_ps_set_elp function and call acx_sleep_auth directly. Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
wl1251: implement filtering Finish up filtering code and enable it. mac80211 no longer requires to be atomic in .configure_filter callback, so remove filter_work. The new filtering code makes sure it doesn't enable BSSID filtering while BSSID is not set so that association works properly. This allows removing hack in wl1251_cmd_join. Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
wl1251: fix channel switching in monitor mode Use the ENABLE_RX command for channel switching when no interface is present (monitor mode only). The advantage of ENABLE_RX is that it leaves the tx data path disabled in firmware, whereas the usual JOIN command seems to transmit some frames at firmware level. Signed-off-by: David Gnedt <david.gnedt@davizone.at>
wl1251: split RX and TX data path initialisation Split up data path initialisation into RX and TX data path initialisation functions. This change is required for channel switching in monitor mode. Signed-off-by: David Gnedt <david.gnedt@davizone.at>
wl1251: fix scan behaviour while not associated With a dissacociated card I often encoutered very long scan delays. My guess is that it has something to do with the cards DTIM handling and another firmware bug mentioned in the TI WLAN driver, which is described as the card may never end scanning if the channel is overloaded because it can't send probe requests. I think the firmware somehow also tries to receive DTIM messages when the BSSID is not set. Therefore most of the time it waits for DTIM messages and can't do scanning work. Anyway we can workaround this misbehaviour by setting the HIGH_PRIORITY bit for scans in disassociated state. [notasas@gmail.com: add missing include and remove debug portion that got comments on l-w] Signed-off-by: David Gnedt <david.gnedt@davizone.at>
wl1251: set rate index and preamble flag on received packets Set the rate index rate_idx and preamble flag RX_FLAG_SHORTPRE on received packets. Signed-off-by: David Gnedt <david.gnedt@davizone.at> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
wl1251: enable beacon early termination while in power-saving mode Port the beacon early termination feature from wl1251 driver version included in the Maemo Fremantle kernel. It is enabled when going to power-saving mode and disabled when leaving power-saving mode. Signed-off-by: David Gnedt <david.gnedt@davizone.at> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
wl1251: fix 4-byte TX buffer alignment This implements TX buffer alignment for cloned or too small skb by copying and replacing the original skb. Recent changes in wireless-testing seems to make this really necessary. Signed-off-by: David Gnedt <david.gnedt@davizone.at> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
wl1251: fix queue stopping/waking for TX path The queue stopping/waking functionality was broken in a way that could cause the TX to stall if the right circumstances are met. The problem was caused by tx_work, which is scheduled on each TX operation. If the firmware buffer is full, tx_work does nothing. In combinition with stopped queues or non-continues transfers, tx_work is never scheduled again. Moreover the low watermark introduced by 9df86e2e702c6d5547aced7f241addd2d698bb11 never takes effect because of some old code. Solve this by scheduling tx_work every time tx_queue is non-empty and firmware buffer is freed on tx_complete. This also solves a possible but unlikely case: If less frames than the high watermark are queued, but more than firmware buffer can hold. This results in queues staying awake but the only scheduled tx_work doesn't transfer all frames, so the remaining frames are stuck in the queue until more frames get queued and tx_work is scheduled again. Signed-off-by: David Gnedt <david.gnedt@davizone.at> Acked-by: Kalle Valo <kvalo@adurom.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>